eBay’s Big Bet on AI

A look at eBay’s enterprise AI system

Add bookmark

One of the first well-known online auction sites in the world, over the past 26 years eBay has achieved long-term success in an environment where that is exceedingly rare. In fact, these days, the average company only survives for about 10 years before it is bought, acquired or liquidated, according to a comprehensive study by the Santa Fe Institute.

However, in the case of eBay, it’s survival can be attributed to two things: agility and data-driven innovation. In fact, over the past twenty years, eBay has emerged as a pioneer of artificial intelligence (AI) and data analytics

 

eBay’s Homegrown, Open Source AI System

In 2018 eBay launched Krylov, a “state-of-the-art AI platform designed to boost eBay’s productivity with AI and accelerate time to market of AI models at scale.” 

To elaborate, Kylov is a multi-tenant, cloud-based enterprise AI platform able to support a diverse set of AI use cases such as merchandising recommendations, buyer personalization, seller price guidance, language translation and shipping estimates. In other worlds, it essentially provides AI infrastructure and frameworks on demand. 

For example, using the eBay app, customers can take a photo of what they are looking for and using computer vision, the platform searches for and identifies items that match the image. 

In addition, Kylov also both increases and democratizes innovation by centralizing collaboration. As Sanjeev Katariya, the Vice President and Chief Architect of the eBay AI and platforms explained in a company blog post, “We had so many engineers and scientists across the company who needed help creating models and pushing out their models to production. We needed a complete closed loop on life cycle management of machine learning algorithms that was obvious. We needed a unified AI platform to really bring data scientists and engineers, modes and management experimentation all together.”

 

Predictive Personalization

One of the primary use cases of machine learning (ML) for any retailer or ecommerce site is personalized recommendations. As for eBay, it launched the latest iteration of its personalization engine back in 2019. 

Designed to enhance the user experience, amongst many other things, this new tool uses predictive analytics to:

  • recommend items based on Recently Viewed Items and past shopping behavior
  • create a the “Trending in Your Interests” feature - a recommendation engine that provides suggestions based on both past shopping behavior and larger, global trends
  • enable the consumer to “Buy Again with One Click” 
  • customize offers, discount and marketing messaging 
  • provide customer with a digital personal shopper
  • allow the user to “Save and Watch” - a feature that alerts customers when specified items hit the marketplace or, if sold out, come back in stock

During an Emerj podcast interview, Zoher Karu, eBay’s Chief Data Officer, elaborated on the company’s approach to personalization, “Some of the ways we’re employing ML and AI to help the buy-in side of things is to understand individual customers’ individual preferences…we have over a 1 billion different items for sale on eBay, and we can use advanced data science techniques to help you shape that list for individual people.

For example, we might try to predict, out of the many different categories of things we sell on eBay, which category are you likely to be most interested in buying next, and we do that by looking at some of your past behavior on eBay, your search behavior, how you’ve interacted with eBay, how other people who are like you have interacted with eBay.”

 

Become a Member of the AI, Data & Analytics Network TODAY!


RECOMMENDED